WesandClaude Sonnet 4.5
27afedb32e
Fetches SMART data from Scrutiny's InfluxDB and calculates failure probabilities using the exact BackBlaze tables from SnapRAID source. Key features: - Calculates both v12 (with attr 193) and v13 (without attr 193) algorithms - v12 matches SnapRAID pre-v13.0 (includes Load Cycle Count) - v13 matches modern SnapRAID v13.0+ (excludes Load Cycle Count) - Stores both values for trending and comparison - Correctly applies bit masks (16-bit for 187/188, 32-bit for others) - Annualizes monthly rates and applies Poisson distribution - Logs to PostgreSQL with device metadata Solved the mystery: SnapRAID v12 uses attribute 193 which was removed in v13.0. Load cycle count has massive impact on failure predictions for some drives (80% vs 4% difference). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
73 lines
2.1 KiB
SQL
73 lines
2.1 KiB
SQL
-- Migration: Add v13.0 failure probability column
|
|
-- Adds a second failure probability column for the modern algorithm (without attr 193)
|
|
|
|
-- Add new column for v13.0+ algorithm (without attribute 193)
|
|
ALTER TABLE smart_metrics
|
|
ADD COLUMN IF NOT EXISTS failure_probability_v13_pct DECIMAL(5, 2);
|
|
|
|
-- Rename existing column to clarify it includes attribute 193 (pre-v13.0)
|
|
ALTER TABLE smart_metrics
|
|
RENAME COLUMN failure_probability_pct TO failure_probability_v12_pct;
|
|
|
|
-- Update views to include both columns
|
|
DROP VIEW IF EXISTS smart_latest CASCADE;
|
|
CREATE VIEW smart_latest AS
|
|
SELECT
|
|
m.id,
|
|
m.timestamp,
|
|
m.device_wwn,
|
|
d.device_path,
|
|
d.serial_number,
|
|
d.model,
|
|
d.manufacturer,
|
|
d.capacity_bytes,
|
|
d.size_tb,
|
|
d.disk_role,
|
|
m.temperature_celsius,
|
|
m.power_on_days,
|
|
m.error_count,
|
|
m.failure_probability_v12_pct, -- With attr 193 (matches your current SnapRAID)
|
|
m.failure_probability_v13_pct, -- Without attr 193 (modern SnapRAID v13.0+)
|
|
m.smart_attributes
|
|
FROM smart_metrics m
|
|
INNER JOIN devices d ON m.device_wwn = d.device_wwn
|
|
WHERE m.timestamp = (
|
|
SELECT MAX(m2.timestamp)
|
|
FROM smart_metrics m2
|
|
WHERE m2.device_wwn = m.device_wwn
|
|
);
|
|
|
|
DROP VIEW IF EXISTS smart_high_risk;
|
|
CREATE VIEW smart_high_risk AS
|
|
SELECT *
|
|
FROM smart_latest
|
|
WHERE failure_probability_v12_pct > 50 -- Using v12 since that matches your SnapRAID
|
|
ORDER BY failure_probability_v12_pct DESC;
|
|
|
|
DROP VIEW IF EXISTS smart_summary;
|
|
CREATE VIEW smart_summary AS
|
|
SELECT
|
|
device_path,
|
|
serial_number,
|
|
temperature_celsius,
|
|
power_on_days,
|
|
error_count,
|
|
failure_probability_v12_pct, -- With attr 193
|
|
failure_probability_v13_pct, -- Without attr 193
|
|
size_tb,
|
|
disk_role
|
|
FROM smart_latest
|
|
ORDER BY
|
|
CASE disk_role
|
|
WHEN 'parity' THEN 1
|
|
WHEN '2-parity' THEN 2
|
|
ELSE 3
|
|
END,
|
|
device_path;
|
|
|
|
COMMENT ON COLUMN smart_metrics.failure_probability_v12_pct IS
|
|
'Failure probability using SnapRAID <v13.0 algorithm (includes attribute 193 Load Cycle Count)';
|
|
|
|
COMMENT ON COLUMN smart_metrics.failure_probability_v13_pct IS
|
|
'Failure probability using SnapRAID v13.0+ algorithm (excludes attribute 193)';
|